Project Summary: Large-scale measurement of individual neuronal activity in intact animals will accelerate the understanding of the brain and treatment of neurological disorders. Fortunately, the last decade has witnessed dramatic improvements in optical methods to record neural activity based on new genetically encoded calcium- or voltage-dependent fluorescence proteins. These optical techniques have the potential to record activity from many thousands of neurons with single-cell resolution because cell activity can be imaged from the surface of the brain without implants that damage neural tissue. Unfortunately, the current tools for fluorescence microscopy in freely moving animals are incapable of recording from more than a few hundred cells at a time due to the large microscope size and small field of view (FOV). To achieve simultaneous imaging of thousands of individual neurons in mammals, fluorescence microscopes must be miniaturized and arrayed so that animals can freely interact with their environment while images of neural activity are constantly recorded over large areas of brain. The goal of our work is to create a new class of flat microscopes (each with a large FOV) that can be arrayed and placed on the brain of a free-moving animal. These microscope arrays will thereby provide continuous imaging over large areas of the brain with cellular resolution in freely moving animals. We also envision that these FlatScopes could be implanted into the brain to measure neural activity from regions that are too deep to image from the surface. To create these FlatScopes we will exploit emerging technologies from the field computational imaging, which make it now possible to replace the expensive, heavy and thick lenses in microscopes with a compact, lightweight, and inexpensive diffractive mask placed near the sensor. Images can then be reconstructed using algorithms that recover the fluorescence images from the multiplexed sensor measurements. Our team of PIs with expertise in computational imaging, nanofabricated neural interfaces, and in vivo neural data acquisition will work to translate the ideas of lens-free imaging to implantable microfabricated FlatScopes that can image neural activity in freely moving animals. Our goal with R21 funding is to design, fabricate, and characterize individual implantable FlatScopes, both in vitro and in vivo, laying the groundwork for a scalable imaging technology to measure calcium- or voltage- sensitive fluorescence in thousands of neurons with single cell resolution.